The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill explores the repository by reading multiple files including CLAUDE.md, AGENTS.md, CONTEXT.md, CONTEXT-MAP.md, and project manifest files like package.json. These files serve as untrusted input sources that influence how the skill scaffolds the issue tracker and domain documentation.
Ingestion points: git remote -v, .git/config, CLAUDE.md, AGENTS.md, CONTEXT.md, CONTEXT-MAP.md, docs/adr/, package.json, and pnpm-workspace.yaml.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the exploration step.
Capability inventory: The skill has capabilities to read files, execute directory/file existence checks, run git commands, create new markdown files in docs/agents/, and update existing CLAUDE.md or AGENTS.md files.
Sanitization: No explicit sanitization of the content read from the repository files is documented before it is used to summarize findings or generate drafts.
[COMMAND_EXECUTION]: The skill instructions and generated templates utilize standard command-line interfaces for repository management and issue tracking.
Evidence: The skill uses git remote -v for environment discovery and provides templates for gh (GitHub CLI) and glab (GitLab CLI) to perform operations such as creating issues, viewing comments, and managing labels.

setup-matt-pocock-skills